Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    a

    np_max-act · gemini-2.0-flash

    The neuron activates on descriptive terms used in personal biographies—i.e. adjectives and role-titles (former, self-taught, dedicated, active, father, etc.) that characterize someone’s background or status.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     مراج
    -0.08
     �
    -0.07
     Gö
    -0.07
     nz
    -0.07
     кал
    -0.06
     confer
    -0.06
     Рас
    -0.06
    ulled
    -0.06
     Pager
    -0.06
     وج
    -0.06
    POSITIVE LOGITS
    .LocalDate
    0.06
     클래스
    0.06
     stiffness
    0.06
     WTF
    0.06
    类
    0.06
     第一
    0.06
     enhanced
    0.06
    นๆ
    0.06
    
    0.06
    >//
    0.06
    Act Density 0.046%

    No Known Activations